Ground Rents Do the risks outweigh the rewards?

Ground Rents Do the risks outweigh the rewards? Strategy & External Affairs November 207

Network Homes: Ground Rents The government s recent consultation on leasehold practices proposes significant changes in the national approach to ground rents. Research we have undertaken at Network Homes leads us to believe the government is broadly taking the right approach, but also that housing associations may well be seriously underestimating some of the risks attached to their ground rent portfolios. Network Homes believes that: Ground rents should be set at less than 0.% of the property valuation and should increase at no faster than inflation. Our research shows that this will ensure ground rents are neither set at, nor increase to, an onerous level Ground rents linked to inflation are likely to be both reasonably fair to the leaseholder and to carry less downside risk for the freeholder Ground rents which double every 25 years (or less frequently) are more likely to track long-term average inflation; those which double every 5 years (or more frequently) are almost certain to exceed inflation and prove onerous Housing associations subject to head leases with onerous ground rent terms should be protected on the same basis as individual leaseholders All housing associations should carry out an investigation of the ground rents they charge and have a clear ground rent policy in place to manage their portfolios effectively For existing leaseholders with onerous ground rents, a voluntary approach should be adopted, where either the freeholder changes the terms of the ground rent or the leaseholder is allowed to extend their lease at reasonable cost and the ground rent reverts to a peppercorn If the government returns ground rents on new leasehold homes to a peppercorn, the development and investment industries will quickly adjust and the impact on new housing supply is likely to be minimal How ground rents work Ground rents fall into three broad categories: those that are fixed, those that increase in line with, and those that double every x years. It is the final one of these categories that can be onerous, however as we explain below, this is dependent on the x, and it is this category that negative media articles on ground rents focus on. An example of such an article was published in The Telegraph in November 206, with the headline Leasehold scandal: ground rent that starts at 250, then rockets to 69 trillion. This article goes on to explain that leaseholders in a block of flats in Islington are being charged an initial ground rent set at 250,doubling every 25 years, with a lease set at 999 years. The This is based on Nationwide s assessment and new approach to issuing mortgages to leasehold properties, http://www.nationwide.co.uk/about/media-centre-and-specialist-areas/media-centre/pressreleases/archive/207/5/05-protect-homeowners

5 0 5 20 25 30 35 40 45 0 20 30 40 50 60 70 80 90 2 23 34 45 56 67 78 89 00 22 26 5 76 0 26 5 76 20 226 Telegraph was wrong in this claim. Firstly, the final ground rent that would be charged annually on such a lease would be 38 trillion! They were also sensationalist as this figure does not mention that this would be the nominal and not the real price, i.e. no adjustment for inflation is made. Indeed, making a moderate assumption of inflation at 2.75% the real ground rent by year 999 will be 240, peaking at 450 per annum in year 976. Our research explored the dynamics of doubling ground rents (in real terms) based on different inflation assumptions and different frequencies of doubling. 400.00 350.00 300.00 250.00 200.00 50.00 00.00 50.00 Graphs of Different Assumptions and Impact on Ground Rents Doubling Every 25 Years 2.5% 2.75% 3% 600.00 500.00 400.00 300.00 200.00 00.00 2.5% 2.75% 3% 0.00 0.00 Graph. Comparison Year -25 Graph 2. Comparison Year -250,200.00 6,000.00,000.00 2.5% 5,000.00 800.00 4,000.00 2.5% 600.00 400.00 200.00 0.00 2.75% 3% 3,000.00 2,000.00,000.00 0.00 2.75 % 3% Graph 3. Comparison Year -500 Graph 4. Comparison Year -999 The graphs above model how much a ground rent starting at 250 per annum doubling every 25 years would cost in real terms, assuming different levels of inflation. Graphs & 2 demonstrate that whilst there is some sensitivity in the first 25-250 years of a lease, the ground rents would not reach a level considered truly onerous. Graph 3 & 4 demonstrate that the true sensitivity arises over the far longer term, with real ground rents reaching close to 5,000 per annum if long term inflation was to be as low as 2.5%. Although this should be somewhat concerning, it would really only impact those who hold a 999 year lease and do not choose to extend their lease.

40 79 8 57 96 235 274 33 352 39 430 469 508 547 586 625 664 703 742 78 820 859 898 937 976 73 45 27 289 36 433 505 577 649 72 793 865 937 68 35 202 269 336 403 470 537 604 67 738 805 872 939 60 9 78 237 296 355 44 473 532 59 650 709 768 827 886 945 73 45 27 289 36 433 505 577 649 72 793 865 937 Ultimately, the sensitivity to inflation should be understood as strengthening the argument for ground rents to increase simply in line with inflation. From the investor side it avoids making a loss in real terms in times of high inflation, and from a leaseholder side it avoids seeing their ground rents sky rocketing in real terms during low inflationary periods. Real Value for Different Frequencies of Doubling Ground Rents Starting at 250 (=2.75%) 300 300 250 250 200 200 50 50 00 00 50 50 0 0 Graph 5. Doubling Every 40 Years Graph 6. Doubling Every 30 Years 500 400 300 200 00 0 Graph 7. Doubling Every 25 Year 500,000 400,000 300,000 200,000 00,000 0 Graph 8. Doubling Every 20 Year 400,000,000,000,000,000,000 300,000,000,000,000,000,000 200,000,000,000,000,000,000 00,000,000,000,000,000,000 0 Graph 9. Doubling Every 0 Year In the graphs above it is worth paying special attention to the values on the vertical axis. Thus, while graph 7 has a an upward trend (this is the graph for the Islington flats), over 999 years it will never exceed 450 per annum (Year 976) in real terms and averages 249.88 in real terms over the course of the 999 years. Graph 8 has an axis with units of 000 s, and peaks at over 400,000 per annum (Year 98) in real terms. However, it is worth noting that if the lease for such a ground rent was 25 years, the maximum ground rent would be 67 per annum. Although this represents more than a doubling in real terms, it is still not on the scale of the figures being discussed in the media. Finally, graph 9 has units of quintillions (0 8 ) and reaches a

32 63 94 25 56 87 28 249 280 3 342 373 404 435 466 497 528 559 590 62 652 683 74 745 776 807 838 869 900 93 962 993 Natural Log of Annual Real Ground Rent staggering peak of 343,476,743,560,45,000,000 (Year 99). In short, the doubling model can cause staggeringly high ground rents, however this will depend on the frequency of doubling, the length of the lease and the rate of inflation. Graph 0 below shows more clearly the dynamics of doubling ground rents, by taking the natural log of the annual ground rent. Every 5 year increase in frequency leads to an approximate doubling in the gradient of the natural log of the real annual ground rent. 50 40 30 20 0 0-0 0 Year 5 Year 20 Year 25 Year 30 Year 35 Year 40 Year Graph 0. Natual Logs of Real Ground Rents Finally, to understand the relationship between our inflation assumptions, inflation sensitivities and the frequency of doubling a ground rent, we have a graph which shows what level inflation would have to be to make the doubling frequency equate to no real change in ground rent levels (graph overleaf). This demonstrates that ground rents that double every 5 years (or more frequently) would need a long term annual inflation rate far higher than anyone could reasonably expect based on historic data. The last century saw median inflation at 3.23%; the last half century 3.60%, but for the last 30 years it was 2.33% 2. Ground rents that double every 20 years, whilst potentially onerous, would only need long term average inflation to reach 3.53% to equate to no change in value in real terms. Although this is not beyond the realms of possibility, and would be quite a fair estimate for, it would certainly be at the higher end of long-term CPI expectations. 2 http://www.bankofengland.co.uk/research/pages/datasets/default.aspx Millennium of Macroeconomic data

8.00% 7.00% 6.00% 5.00% 4.00% 3.00% 2.00%.00% 0.00% 7.8% 4.73% 3.53% 2.8% 2.34% 0 5 20 25 30 35 40 Graph. Frequency Doubling and Required Inflation Rate for Real Value to Not Change 2%.75% In short, once we take account of inflation it is fair to assume that ground rents that double every 25 years (or less frequently) will more than likely track long-term average inflation (or underperform), and those doubling every 5 years (or more frequently) are almost certain to far exceed inflation and become onerous. This demonstrates that some of the horror stories are misrepresentative of the true state of affairs, but at the same time there are genuinely onerous and exorbitant ground rents being set. Furthermore, from a developer or investor side the research demonstrates that ground rents that double every x years are very high risk and do not guarantee inflation proof returns, and their real rate of returns are very sensitive to long-term average inflation rates. There appears to be a clear solution that will satisfy both parties, namely, establishing a status quo whereby ground rents increase in line with inflation (whether this should be, CPI, or CPIH is beyond the scope of this paper). The issues facing housing associations As outlined above, ground rents can increase in a wide number of ways. Some of these can lead to onerous annual charges for the leaseholder, while others can constitute a legitimate and valuable income stream to the freeholder. Housing associations may find themselves on both sides of the argument therefore. Firstly, many housing associations are freeholders with considerable ground rents portfolios, both with regards to number of leases and total value of collectable ground rent. However, due to the history of housing associations, with many having built up over many years, often including stock transfers, mergers of other entities and developing mixed tenures, the number of ground rent types (i.e. the terms and the frequency of increase) range widely. This creates a particular challenge for housing associations, both with regards to the efficient and effective management of ground rents as they currently stand, and in ascertaining the financial impact any retrospective changes may have on their finances moving forward. All housing associations should carry out a full investigation of the ground rents they charge, if they have not done so already.

Secondly, many housing associations are leaseholders themselves, specifically holding Head Leases, with under-leases or tenanted properties held within the Head Lease. The reason for the prevalence of Head Leases is that for more than 20 years Section 06 provisions have been one of the primary mechanisms by which social housing has been delivered. Thus, in the case of a block of flats, it is not uncommon for the developer to retain freehold interest, with housing associations obtaining the Head Lease containing the social housing units. Such Head Leases may contain ground rents as well. These in turn can be just as onerous as the ground rents discussed before. Where there are under-leases (leasehold or shared-owner) the ground rent liability may pass on through to the leaseholders, or where there are tenanted properties the housing association may be solely liable. This creates a dual challenge. Firstly, in the case of under-leases where onerous ground rents are dictated in the Head Lease, the housing association may wish to change all of their ground rents to be non-onerous, in which case if no provision is made to ensure the Head Lease is amended, the housing association would be liable to cover the shortfall. Furthermore, in the case of tenanted properties, where the Head Lease includes onerous ground rents, the housing association may already be liable to paying these onerous charges, and may have no recourse to ensure the freeholder amends the ground rent to a reasonable level. Thus, we urge the government to ensure that if provisions are made to protect leaseholders from onerous ground rents (and/or allow for retrospective amendments), these should also be applied to Head Leases held by housing associations. Not only is this a matter of protecting housing associations from unfair charges, which in turn may undermine their ability to deliver more social housing, this is also a matter of principle. The provision of social housing via Section 06 and the use of Head Leases to transfer properties from private developers to housing associations was surely never intended to be an opportunity for the private developer to make additional profit; rather it was intended to ensure that private developers were contributing to the local community. The instances in which developers have set ground rents at onerous levels in Head Leases could be seen as private developers acting contrary to the intended spirit of the legislation. A contemporary history of ground rents Given the risk of ground rents increasing to such astronomical levels, we must ask ourselves: how did we get here? For most of the 20 th century ground rents were set at peppercorn amounts, typically 0 per annum with no increases. However, at some point in the last 20-30 years developers and subsequently investors identified ground rents as a means of maximising income from their properties. At this point, ground rents went from peppercorn amounts (or close to it), to being set at higher levels (typically 50-400 per annum) and more crucially had regular increases prescribed in the contract. The terms of the increase varied, falling into the two main categories discussed above: those that increase in line with (every 5 years or 0 years), and those that double every x years. To understand how the more onerous ground rents

grew in popularity amongst developers and investors, we carried out a review of industry articles relating to ground rents over time. In 200 both Savills and CBRE published reports on ground rents as a good investment asset this can be said to be the watershed moment for ground rents establishing themselves in the mainstream. It is worth noting that CBRE s report in 200 identified ground rents that increase in line with as being the most investor friendly form. Their reasoning was that investors were seeking ground rents as a low risk but inflation proof investment; moreover CBRE identified the risk that ground rents that increased too aggressively could ultimately undermine their own low risk profile. The low risk nature derives from the fact that if a leaseholder fails to pay their ground rent they risk forfeiting their property; where ground rents are a small proportion of the value of the property the risk of failure to pay is close to zero, however if ground rents are set at (or increase to) a higher proportion of the property s value, the risk of non-payment will increase. From 200 to the end of 205 there was an increasing number of industry reports analysing the ground rent market, and by the end of 205 CBRE published a report highlighting the rapid pace in market growth, with 0 transactions exceeding 0m in the preceding 8 months. However, the question must be asked: why was there such a large amount of investment in what is a rather dull asset? The consensus is that in the wake of the global recession investors, especially large institutional investors, sought low risk inflation-proof assets and ground rents, at least on paper, appeared to fit the profile. On the developer side, ground rents offered an opportunity to raise much needed additional capital effectively out of thin air. Given investors were primarily attracted by their inflation proof low-risk returns, how did it come to pass that freeholders started setting ground rents with onerous increases? With certain developers assuming increased income from ground rents, other developers will have been forced to make similar assumptions to remain competitive. In under-regulated markets, unacceptable but profit-maximising practices can spread quickly. It is clear that as ground rents went from simply increasing in line with inflation to doubling at ever more frequent intervals, their fundamental investment profile shifted from being low-risk and inflation proof, to ever more risky. However, once these onerous ground rents were more widespread, where were the valuers, mortgage providers and solicitors, all in theory meant to be acting in the would-be leaseholders interests. In short, why was there no professional push-back sooner? The primary motive of the developers who set the ground rents is to maximise profits. However, we believe developers/investors have underestimated the risk with ground rents. Specifically, the additional risk posed to ground rents as an investment is that of an imminent increase in interest rates. A ground rent portfolio s valuation will be negatively correlated to interest rates, perhaps more so than many realise. As interest rates increase, investors will increasingly move to holding cash as a low risk option, but when rates are low they will seek out alternative low-risk assets such as ground rent portfolios. The second manner, which appears to be ignored, is that when interest rates increase, property prices are likely to decrease, and as this happens

ground rents will become riskier as the ground rent will increase as a proportion of the property value. This, in turn, will increase the likelihood of failure to pay. Overall, while the last decade has seen ground rent investment go from a fringe product to the mainstream it is safe to say that the last 2 months has taken its toll on the ground rent market. A brief look at the Ground Rents Income Fund Plc, a fund specialising in ground rent investment, shows that the increased government focus on the issue has hit their share price. After the government consultation on leasehold reform was announced the on 25 th July, their share price fell by 4%, eventually falling by a total of % by st August. Money for nothing? The history of ground rents, along with leasehold as a tenure, dates back to the th century, thus to describe them as feudal is factual. While we may say they are an annual payment from a leaseholder to the freeholder, what is a ground rent a payment for? Freeholders can recoup from leaseholders a range of costs they may incur in the form of service charges, major works costs and management fees. All of these fees leaseholders are charged relate directly to a cost incurred by the freeholder. Ground rents are unique in this respect in that they relate to no cost incurred by the freeholder. It is an archetypal economic rent: the part of the produce that accrues to the owners of land by virtue of ownership (Henry George, 879). It is money paid by the leaseholder to the freeholder for nothing, other than the simple fact that the freeholder is the ultimate owner 3. The question arises, given their archaic and arguably unjustifiable nature, why do ground rents still exist? The best explanation is that for most of the modern period ground rents were little more than nominal fees which had no terms of increasing; so small that many freeholders did not bother to collect them. They were included in leasehold contracts as a vestigial organ, serving little function, and were simply a reminder of the evolutionary past of leasehold. If ground rents serve only as an economic rent, the government s intention to reform the market and return them to a peppercorn level is fair. The development market will quickly readjust its pricing to take account of this and there is unlikely to be more than a minimal impact on new housing supply. Conclusion Network Homes supports tighter regulation of the setting of ground rents to ensure they are not onerous, which in turn will not only protect leaseholders, but also ensure leasehold properties are still eligible for mortgages (which will support developers and potential buyers), and finally will minimise the risk to investors. We believe that less than 0.% of the property value adjusted in line with will satisfy the interest of the various parties involved. 3 The argument that ground rent is compensation for some supposed opportunity cost, simply raises the question of what the initial cost of the lease relates to.

For pre-existing leaseholders, we believe that initially solutions should be sought on a voluntary basis via amendments to terms in leases, or allowing for leases to be extended for a nonexorbitant cost. This allows for a more pragmatic and flexible approach to resolving these problems. However, where no voluntary solution can be found we would not oppose government intervention. Given the role head leases play in the provision of section 06 properties, a mechanism intended to increase affordable housing and not maximise profits for private developers, we would like the government to also address the ground rents on head leases. We would support these ground rents being subject to the same regulation as for the leases of owner-occupiers, and that any retrospective action also be applied to said head leases. Finally, given the feudalistic history and lack of clear purpose in the modern era, we would encourage the wider house-building community to meditate on the question of whether or not ground rents should be part of a more transparent and fairer future? Kristian Scrase Research & Intelligence Analyst